© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 6631
    Prev
    Next
    INDEX
    Explanations

    words or phrases related to reassurance or advice

    oai_token-act-pair · gpt-3.5-turbo

    repeated instances of the word "worry" and variations of it

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ewitness
    -1.00
    arb
    -0.91
    ĻĤ
    -0.76
    artifacts
    -0.75
    assisted
    -0.71
    avorite
    -0.70
    authorized
    -0.69
    pled
    -0.69
     authorized
    -0.67
    endum
    -0.67
    POSITIVE LOGITS
    warts
    0.97
    lessly
    0.94
     worry
    0.88
    wart
    0.85
     fret
    0.82
    ingly
    0.82
     worrying
    0.76
     Pu
    0.76
    fulness
    0.75
     ABOUT
    0.75
    Activations Density 0.020%

    No Known Activations